Computing a partial Schur factorization of nonlinear eigenvalue problems using the infinite Arnoldi method (Q2923350)
From MaRDI portal
| This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use this page instead for the normal view: Computing a partial Schur factorization of nonlinear eigenvalue problems using the infinite Arnoldi method |
scientific article; zbMATH DE number 6356178
| Language | Label | Description | Also known as |
|---|---|---|---|
| English | Computing a partial Schur factorization of nonlinear eigenvalue problems using the infinite Arnoldi method |
scientific article; zbMATH DE number 6356178 |
Statements
15 October 2014
0 references
Arnoldi's method
0 references
nonlinear eigenvalue problems
0 references
invariant pairs
0 references
restarting
0 references
partial Schur factorization
0 references
numerical example
0 references
Computing a partial Schur factorization of nonlinear eigenvalue problems using the infinite Arnoldi method (English)
0 references
The authors consider the nonlinear eigenvalue problem (NEP) of finding \(\lambda \in \Omega\subseteq\mathbb{C}\) and \(v\in\mathbb{C}^n\backslash\{0\}\) such that \(M(\lambda)v=0\), where \(M:\Omega\leftarrow\mathbb{C}^{n\times n}\) is analytic in \(\Omega\), which is an open disc centered at the origin. They introduce a new technique to compute a partial Schur factorization of the NEP based on the infinite Arnoldi method (cf. [\textit{E. Jarlebring} et al., Numer. Math. 122, No. 1, 169--195 (2012; Zbl 1256.65043)]). The modification applies the fact that the invariant pairs of the operator are equivalent to invariant pairs of the NEP. Using the characterization of the structure of the invariant pairs of the operator, they show how to modify the infinite Arnoldi method by respecting this structure. Finally, they present two numerical examples to justify the usability of the main algorithm.
0 references